Etherscan Incorporates ChatGPT to Analyze Ethereum Source Code

Updated: 20 June 2023

Disclosure

Don’t invest unless prepared to lose all the money you invest. This is a high-risk investment, you shouldn’t expect to be protected if something goes wrong.

Join Our Telegram channel to stay up to date on breaking news coverage

Etherscan, one of the leading Ethereum blockchain explorers, has announced its integration with ChatGPT to enhance the analysis of Ethereum source code. With this new feature, developers can now have a better understanding of smart contract code and reduce the chance of potential errors.

With the basic artificial intelligence functionality of the service, users can now retrieve and interpret the source code for a specific contract address.

The new tool incorporates artificial intelligence (AI) into the capabilities of Etherscan’s blockchain explorer. As per a tweet by Etherscan the beta version of Code Reader “leverage the power of AI to seamlessly learn about any smart contract source code!”

🆕 Code Reader (Beta)

Leverage the power of AI to seamlessly learn about any smart contract source code! ✨ pic.twitter.com/GTbULisudk

— Etherscan (@etherscan) June 19, 2023

Since its built on AI technology created by OpenAI, the tool enables users to request an explanation of all or a portion of a smart contract’s source code.

By providing access to a smart contract’s “read” and “write” capabilities, users can make informed decisions about how to engage with smart contracts and investigate their potential uses in decentralized apps.

Etherscan AI-Powered Code Reader

Following the significant growth in AI interest, kindled by the impressive success of OpenAI’s ChatGPT chatbot, various blockchain and cryptocurrency companies have been quick to offer traders and developers solutions based on this technology.

The AI-powered tool from Etherscan comes after the just-debuted filter tool, unveiled on June 2, 2023. However, by leveraging AI these tools can improve the data for researchers, on-chain detectives, and investigators. Beginning in 2023, AI-related blockchain and cryptocurrency concepts have begun to take shape.

A few examples of these ideas are tokens with an AI focus and chatbots that are knowledgeable about cryptocurrencies like Bitcoin (BTC).

Etherscan claims that the tool is only meant to be used for educational purposes. Additionally Etherscan cautions users from relying only on the tool’s findings in litigation or bug bounty applications.

To use Code Reader, users must connect to the OpenAI API and have the appropriate consumption limits. An API, or application programming interface, facilitates information sharing and communication between two computer systems. However, currently, Etherscan’s technology doesn’t support discussion threads, so the chatbot can only be accessed through one-time requests.

“By selecting a contract address, users can access and interact with the contract’s source code files,” as noted in the Code Reader walkthrough article. The tool leverages Openai’s large language model (LLM) to accelerate intelligent contract research and development.

🆕 Code Reader (Beta)

Leverage the power of AI to seamlessly learn about any smart contract source code! ✨ pic.twitter.com/GTbULisudk

— Etherscan (@etherscan) June 19, 2023

The Use of ChatGPT

Etherscan is not the only company using ChatGPT. Other businesses are also using ChatGPT to explore blockchains. Alchemy, a renowned blockchain platform developer, recently released their AlchemyAI ChatGPT-based tool, which includes a GPT-4 plugin for blockchain research. Moreover, Solana Labs also have a ChatGPT plugin, which has been available since May.

Despite ChatGPT’s inclusion in blockchain analysis, Etherscan points out the program’s drawbacks. The company advises customers to independently verify the facts and not rely exclusively on ChatGPT’s reply.

This note of caution relates to the pervasive issue of AI chatbots producing erroneous and deceptive information when reacting to instructions (hallucinations).

AI hallucinations refer to situations where an AI generates false conclusions that are not supported by actual data. They may contain erroneous information, news, or details about people, places, or things.

However, OpenAI announced in May that they are implementing new training models to reduce these hallucinations.